Dictionaries and usage guides are necessary tools for editors. Problems arise, however, when the guides and dictionaries disagree or when they say “yes, but.” Such is the case with alright and all right.
Authors, including such notables as Flannery O’Connor, Langston Hughes, Gertrude Stein, Theodore Dreiser, and James Joyce, have used alright, but the consensus seems to be that alright is not all right to use — it is nonstandard English.
That alright is considered substandard English is odd considering fusions of all ready to already and all together to altogether are accepted uncritically. But that is one of the mysteries and beauties of English — the lack of rhyme or reason for something to be okay or not. One theory, advanced by the The American Heritage Guide to Contemporary Usage and Style, is that already and altogether became single words in the Middle Ages, thus before the arrival of the language critic, whereas alright has been around for little more than 100 years (since near the end of the 19th century-beginning of the 20th century), giving language critics an opportunity to cast aspersions on its use.
Even though the words are not always synonymous, some critics, such as Bryan Garner, ignore the differences. As the American Heritage Guide notes, “The sentence The figures are all right means that the figures are all accurate, that is, perfectly correct, while The figures are alright means that they are satisfactory.…”
Merriam Webster’s Dictionary of English Usage asks “Is alright all right?” and answers with a qualified yes: First, all right is more commonly used in print. Second, the authors of most handbooks for writers think alright is wrong. And third, alright is more likely to be found in trade journals, magazines, and newspapers than in more literary sources. (Is word snobbery at play here?)
The earliest use of alright in modern usage is by Chaucer in 1385. But once we leave Chaucer, there are no examples of either alright or all right until the late 17th-early 18th centuries when there are examples of all right but with all used as a pronoun, as, for example, in Defoe’s Robinson Crusoe (1719): “desir’d him to…keep all right in the Ship.”
The first uses of all right as a fixed phrase appear in the early 19th century, as in Shelly’s (1822) Scenes from Goethe’s Faust (“That was all right, my friend.”) and in Dickens’ (1837) Pickwick Papers (“‘All right, sir,’ replied Mr. Weller.”). The first recorded use of alright in modern times was in 1893 in the Durham University Journal.
The controversy over the correctness of alright seems to have begun in the early 20th century. Frank Vizetelly denounced the use of alright in his 1906 book, A Desk Book of Errors in English. In 1924, the Society for Pure English published a symposium on alright by H.W. Fowler of Fowler’s Modern English Usage fame. Fowler considered the word bad spelling and in his 1926 Modern English Usage, he repeated his earlier denunciation of the word. In Fowler’s third edition (R.W. Burchfield, Ed., 1996), the discussion opens with “The use of all right, or inability to see that there is anything wrong with alright, reveals one’s background, upbringing, education, etc., perhaps as much as any word in the language.” The entry concludes, “The sociological divide commands attention.” Basically, Fowler, a word and social snob preferred all right because the hoi polloi prefer alright, an attitude continued by Burchfield. Clearly, a well-reasoned and justifiable position.
According to Merriam Webster’s Dictionary of English Usage, Theodore Dreiser repeatedly used alright in his manuscript but H.L. Mencken, his editor, had him change it to all right. It seems to be a battle between writers (alright) and self-proclaimed language experts (all right). Merriam Webster goes on to say that “undoubtedly [alright] would be even more frequent in print than it is if copy editors were less hostile.” (Editors do have some influence!)
According to Bryan Garner, today’s usage guru, “the combined version [alright] cannot yet be considered good usage — or even colloquially all right” (Garner’s Modern American Usage, 2009). Garner labels alright as a stage 2 word, that is, “widely shunned” on his Language-Change Index. Garner also calls alright an “invariably inferior” word, but without saying why it is “invariably inferior.”
I know that my opinion regarding usage isn’t at the level of esteemed, but this seems to me to be much ado about nothing. Using Garner’s own statement that “the combined version [alright] cannot yet be considered good usage — or even colloquially all right” demonstrates the utility of distinguishing between all right and alright, with both being acceptable when appropriately used. If he had written instead, “the combined version [alright] cannot yet be considered good usage — or even colloquially alright,” it would be clear what “colloquially alright” means. By using all right, it isn’t clear whether alright is colloquially inaccurate or simply unsatisfactory, although we can guess the former from the tenor of his comments. However, if we accept Garner’s statements that the function of language is to communicate clearly, it seems to me that it is perfectly alright to distinguish between all right and alright solely by intended meaning and not by whether some critic thinks one is a better spelling or form than another. It also seems to me that it is all right to always use alright.